Fish

It all starts with fish.

Fish are where we find the beginnings of the mammalian ear, which is much more complicated and we will get to it later. Fish and mammals have one important thing in common, and that’s the inner ear. The inner ear is what transduces the pressure signals from the environment into what we perceive as sound. For humans, we funnel sound into our inner ear through our pinnae, what we see as our external ears. Fish don’t have external ears, because their bodies are just as dense as the water around them, and so sound can pass right through their skin, into their inner ear (Putland, R. L., Montgomery, J. C., & Radford, C. A., 2019). The hair cells in the inner ear that detect the particle movement of sound are denser than the fish, so they don’t actually move the same way as the fish and water do. This displacement is how the fish detects pressure changes, and thus sound. The direction of this displacement also tells the fish where the sound is coming from, which is called localization.

Unfortunately, fish can’t hear that well with just the inner ear, so some fish also have a swim bladder, which is a pocket of gas inside them that is much less dense than their body and water, so it's more sensitive to pressure. The swim bladder connects to the inner ear, providing more information. This typically extends the hearing range of the fish up to around 5 kHz (Putland, R. L., Montgomery, J. C., & Radford, C. A., 2019).

A final mechanism that helps fish hear is called the lateral line. It’s a tube of fluid right below the skin’s surface. Pores connect that tube to the sea water, and thus couple the pressure from the sea water to the later line. Then hair cells inside the lateral line detect pressure changes. This is actually part of how fish swim so well together in schools (Putland, R. L., Montgomery, J. C., & Radford, C. A., 2019). As soon as one fish moves, it creates a pressure change around them, which then makes the other fish move, creating a domino effect so the whole school follows the same path. Of course, vision also plays into this too.

Since these kinds of fish can only hear up to 5 kHz, they also only communicate in this lower spectrum. One highly studied group of fish, the Batrachoididae (toadfishes and midshipmen), make 40-250 Hz boat whistles and 90–140 Hz hums to attract females (Putland, R. L., Montgomery, J. C., & Radford, C. A., 2019). We’ll see later with marine mammals a trend of higher frequency vocalization and hearing, which is a general trend with mammals.

Birds and Reptiles

The next step to get the mammalian ear is with reptiles and birds. When fish left the water, their ears needed to adapt to hearing in the air. This required a new mechanism that coupled pressure changes in the air to pressure changes in the liquid-filled inner ear, as well as another mechanism that allowed the sound to enter the body and travel to the inner coupling device: the middle ear. This is where the middle/external ear is born. Air from the outside travels through a hole—usually in the animal’s head—and vibrates the tympanic membrane, which then transfers that signal along bones in the middle ear called ossicles. These amplify the signal so the fluid in the inner ear, which is more dense than air, can accurately detect sound.

Reptiles and birds only have one ossicle, the columella (Anthwal, N., Joshi, L., & Tucker, A. S., 2013). This is different from mammals, which have three. Only having one ossicle means the hearing range of reptiles and birds is limited to around 8-9 kHz, with best hearing around 250-500 Hz (Dooling, R. J., Lohr, B., & Dent, M. L., 2000). As for the external ear, most reptiles have the tympanic membrane either flush with the skin, or slightly embedded, and don’t have any pinnae (Heffner, H. E., & Heffner, R. S., 2018). Birds are the same story, except some of them do have small pinnae like some owls.

In order to identify the direction of incoming sound, birds and reptiles have to find a different method than fish. This is because the sound will always be coming into their ears in the same direction, through their ear holes and into the tympanic membrane—which is not always the same direction as where the source is. For birds and reptiles—and later mammals—to localize sound in the horizontal plane, the most important cues are: interaural time difference (ITD) and interaural level difference (ILD) (Grothe, B., & Pecka, M., 2014). ITD is the difference in time between when a sound enters one ear versus the other. And ILD is the difference in the sound intensity in one ear versus the other.

The usefulness of ITD and ILD cues is frequency dependent. ITDs are more pronounced for lower frequencies because they travel slower, whereas ILDs are more pronounced for higher frequencies because they get attenuated by the head before reaching the other ear; this is known as the head shadowing effect. Head size and hearing range are an important factor when deciding if ITDs or ILDs are more useful. With a larger head, ITDs will become more pronounced—and thus easier to use—because sound needs to travel further to get from one ear to the other. With a higher frequency range, ILDs are more useful because you can hear more of the head shadowing effect.

Since reptiles and birds can only hear up to 10 kHz, they take more advantage of ITDs. To further pronounce these ITDs, they also have an extra connection between the ears inside the body to couple them. For amphibians and most reptiles, sound travels between the ears through the mouth (Heffner, H. E., & Heffner, R. S., 2018). For crocodiles and birds, sound travels through a different tube, the interaura canal. Connecting the ears turns them into pressure difference receivers, which—due to the fact that sounds can now interfere with each other between the middle ears—actually increases ITDs (Grothe, B., & Pecka, M., 2014). This is especially helpful for birds and reptiles with smaller heads, where ITDs would become almost impossible to hear without this coupling.

But what about the vertical plane? Mammals take advantage of the interaction between high frequencies and pinnae, but reptiles and birds can neither hear those frequencies or have pinnae. For birds, the shape of their head allows them to localize via ILDs of higher frequencies—around 3.5-5.5 kHz (Schnyder, H. A., Vanderelst, D., Bartenstein, S., Firzlaff, U., & Luksch, H., 2014). A spherical shape creates an acoustic shadow, but also diffracts sound towards the other side where it all adds up. Being relatively spherical, the heads of birds do the same thing. By detecting where the shadow is versus the bright spot, birds can tell the elevation of a sound. This occurs contralateral to the source—i.e. on the opposite side of the listener—which is beneficial to birds with lateral eye positions, eyes facing to the left and right; the lateral nature of this effect aligns with their eyesight. For birds with forward facing eyes, typically predators like the barn owl, it would be very confusing, so they have evolved other ways to localize vertically.

The barn owl has two evolutions that help them localize vertically. The first is their feather ruff, which creates a shadowing effect similar to head shadow except suited for vertical localization. The other is an offset of their ear holes such that one sits a bit above the other (Grothe, B., 2018). This allows them to also localize sounds in the vertical plane using the same cues they use in the horizontal plane. Owls in particular have superb audio processing, and can use interaural level and time difference cues across their whole range, which is up to around 9 kHz (Grothe, B., 2018). These cues are plenty for them to precisely localize prey below them. Owls hunt small quick rodents that scuttle on the ground while they fly above, and need to know where to swoop down to catch them before they run away.

Terrestrial Mammals

The driving difference in Mammalian ears is that they are not coupled. Though technically they are both connected to the mouth via the eustachian tube, this connection doesn’t provide enough air travel to help with localization. Its primary function is for regulating the pressure in your middle ears. The reason for the isolation of the two ears is speculated among researchers, but one hypothesis is that mammals are continuous breathers, and so could not have the ears coupled via the mouth or else the sound from breathing would mask external sounds (Heffner, H. E., & Heffner, R. S., 2018). On the other hand, Amphibians and reptiles breathe intermittently—usually hours apart—so they can deal with the interference every now and then. Birds, who are also continuous breathers like mammals, evolved the interaural canal separate from the respiratory system to avoid interference and reap the increased ITDs of coupled ears. There are a few holes in this however. Why do crocodiles also have an interaural canal despite being intermittent breathers? And why did mammals not follow birds and have both continuous breathing and an interaural canal? As for the latter question, it may just not have been important to mammals to have a boost in ITDs given they had much stronger ILD cues because of high frequency hearing. But in general, these questions have yet to be answered.

So how do mammals localize then? Instead of one ossicle, the mammalian middle ear has three: the malleus, incus, and stapes. The stapes is analogous to columella, but where did the malleus and incus bones come from? Well, they came from the jaws of reptiles and birds (Anthwal, N., Joshi, L., & Tucker, A. S., 2013). During this evolution, the whole face and skull was reconstructed as the two bones moved up into the ears. That also meant a new mechanism had to be implemented to move the jaw. Once the malleus and incus were in place, the range of mammalian hearing skyrocketed. As an example, bats are most sensitive to frequencies between 10 and 70 kHz according to Encyclopedia Britannica (retrieved 2023). To put that in perspective, their most sensitive hearing range starts where the range of reptiles and birds end.

The increase in range in mammalian hearing is due to the addition of the malleus and incus, which help the middle ear better transfer the energy from the air to the fluid of the cochlea (The ossicles and their function. (n.d.), retrieved 2023). They act as an impedance-matching mechanism, and amplify the signals so higher frequencies, the frequencies with less energy, can be heard. Being able to hear higher frequencies means mammals take great advantage of interaural level differences (ILDs) to localize. The head shadowing effect works for the horizontal plane, but what about the vertical plane? This is where the external ear, the pinna, comes into play. The pinna helps mammals localize sound in the vertical plane, as high frequencies will interact with the folds of the pinna, changing the spectrum of the sound depending on its elevation relative to the ear (Heffner, H. E., & Heffner, R. S., 2018).

General hearing range of mammals varies wildly depending on the size of the animal. For smaller mammals, their ears are closer together, and their ossicles are lighter, leading to a build best suited for high frequency hearing (Heffner, H. E., & Heffner, R. S., 2018). Since their head is smaller, in order to take advantage of ILDs they need to hear higher frequencies. And since their ossicles are smaller, they would naturally hear higher frequencies anyways. This is just a rule of thumb, as hearing does depend a lot on what information is actually useful to the animal. For example, the naked mole rat is a subterranean rodent. The subterranean soundscape is characterized greatly by low frequency sounds, so it's beneficial for the mole rat to specialize in low frequency hearing (Mason, M. J., Cornwall, H. L., & Smith, E. S. J., 2016). This does mean they cannot use any sound localization cues, but they make up for it using touch. Their hearing range is 65-12.8k Hz. Compare that to the hearing range of a mouse, another small rodent, which is 1-100 kHz (Reynolds, R. P., Kinard, W. L., Degraff, J. J., Leverage, N., & Norton, J. N., 2010).

As for larger mammals, their larger head and ossicles mean they tend towards a lower range of hearing (Heffner, H. E., & Heffner, R. S., 2018). In localizing, these animals thus specialize more in ITDs than ILDs. The elephant is a good example of this. With a hearing range of 17 Hz to 10.5 kHz, they specialize in communicating through infrasounds, sounds less than 20 Hz, below the human hearing threshold. This was beneficial to elephants because they used to live in dense forests, where frequencies as low as these could pass right through the foliage (Garstang, M., 2004). Though they can only locate sounds below 4 kHz, this is plenty for all of their communication needs. Their calls are primarily used for communication regarding reproduction, finding food/water, and avoiding predators. And elephants can tell how far away a low frequency call was made by how much higher frequency content is still left in the signal.

Marine Mammals

After looking at the progress from fish hearing to mammalian hearing, the story seems to be relatively coherently structured around the outer/middle/inner ear paradigm. That's because there are a lot of exceptions to what has been described, namely in insects and some frogs and other reptiles, who hear via other methods (Krause, 2013). The last notable thing about the evolution of the ear regards marine mammals. To recap a bit, the general path of evolution so far has progressed from water to air. Reptiles, birds, and mammals all had to deal with the fact that they got their hearing from fish, who’s inner ears evolved to handle hearing in the ocean. As explained, this is where the outer and middle ears came from—out of a need to couple the vibrations in the air to vibrations in the liquid-filled inner ear. But when marine/amphibious mammals decided to go back into the water, what became of the ear?

All marine mammals retain a liquid-filled inner ear and an air-filled middle ear (​​Reynolds, J. E., & Rommel, S. A. (Eds.), 1999). Otherwise, the differences between the ear of a marine mammal species and a land mammal depend on how well that marine species has adapted to water.

The ears of seals, which are amphibious, are a compromise. Some seals are better adapted to the water, and thus have better hearing in water than on land. For seals like the eared seal—named for the presence of small pinnae—it’s the opposite. But other mammals are heavily adapted to the water, like odontocetes and mysticetes.

Mysticetes are whales which have baleen filtering hairs in their mouth to catch plankton from the water as opposed to odontocetes, which are toothed whales like dolphins and orcas. Mysticetes are like the elephants of the sea, making infrasonic calls to communicate over very long distances. This whale singing helps them locate good places to eat and reproduce. These kinds of whales are migratory, and travel very long distances, so it makes sense as to why low-frequency communication is used.

Odontocetes on the other hand, are sort of like the bats of the ocean, using ultrasound to echolocate. Reynolds and Rommel (1999) put odontocetes into two types based on their vocalization/hearing. Type 1 odontocetes produce/hear sounds that peak about 100 kHz. These are dolphins that live near the shore or in rivers. These kinds of waters are acoustically complex, and thus require higher echolocation precision from higher frequencies. For example, the Amazonian Boutu hunts small fish among roots and stems using echolocation of up to 200 kHz. Communication signals are rarely observed in type 1 odontocetes.

Type 2 odontocetes produce/hear sounds that peak below 80 kHz (​​Reynolds, J. E., & Rommel, S. A. (Eds.), 1999). They live either nearshore, or entirely offshore. These waters are less acoustically dense, so they don’t require as precise echolocation. Type 2 species also travel in groups, devoting more effort to acoustic communication rather than echolocation, another reason why they vocalize lower.

The structure of odontocetes and mysticetes ears is a departure from that of terrestrial mammals. While they do have ear holes, those canals are plugged with wax and also don’t connect to the middle ears (​​Reynolds, J. E., & Rommel, S. A. (Eds.), 1999). These marine mammals have returned to the way fish ears work. Their skin tissue density matches closely to that of sea water, but it isn’t a perfect match. Not much is known about how sound enters mysticetes heads, but for dolphins, some parts of the head are more sensitive to sound than others (Nachtigall, P. E., Lemonds, D. W., & Roitblat, H. L., 2000). There exists a lining of fat along the lower jaw that matches very well with the impedance of seawater. Perpendicular to that is another fatty tissue of the same density on the pan bone, which is positioned near the back of the jaw, above the lower jaw. A third fatty area exists near the melon, or the forehead, of the dolphin (Nachtigall, P. E., 2016). These fatty areas all connect to the middle ear, and are the most sensitive areas for sound conduction.

These fatty areas are key to how odontocetes localize sound (Nachtigall, P. E., Lemonds, D. W., & Roitblat, H. L., 2000). Each point of entry and channel to the middle ear introduces its own attenuation and delay. Since all of these points of entry are located on different places on the vertical plane of the dolphin, any sound heard by the dolphin has had its spectrum modified by such channels depending on the direction of the incoming sound. This is similar to how the folds in the pinna change the spectral characteristics of a sound depending on direction, and is why some researchers have argued that dolphins have inner pinnae. Thus, dolphins can localize really well on the vertical plane.

As for the horizontal plane, it is initially easy to dismiss the dolphin’s ability to localize because of a few reasons. Firstly, a lot of the dolphin’s body matches impedance better with seawater than a land mammal’s body would match impedance with air (Nachtigall, P. E., Lemonds, D. W., & Roitblat, H. L., 2000). Thus it would be easy to assume the dolphin’s ears are not as isolated as they really are, and so cannot take advantage of the same interaural time/level cues land mammals use. This is not true, however, as odontocetes' inner ears are isolated from the skull (Nachtigall, P. E., Lemonds, D. W., & Roitblat, H. L., 2000), and odontocetes can take advantage of ILDs because of head shadowing and ‘inner pinnae.’

As for ITDs, in the ocean, the speed of sound is much higher than it is in air, and so is the wavelength. This makes it more difficult to use the interaural time difference to localize sound for higher frequencies. In addition, since the sound travels directly inside the dolphin, the distance between the cochlea—not head size—is the measurement for ITD calculations. The distance between the cochlea is smaller than their head width, which means it's even harder for them to hear ITDs. Despite this, dolphins have superb processing abilities, and can hear ITDs just as well as a land mammal with similar dimensions can (​​Reynolds, J. E., & Rommel, S. A. (Eds.), 1999). But as discussed earlier, dolphins also take huge advantage of ILDs and spectral cues from their ultra sonic hearing to supplement this.

Conclusion

The history and evolution of the mammalian ear tells a very interesting story about how different animals perceive and locate sounds in the world around them. This paper has taken a look at many different animals and how they take the same data and shape it to suit their needs. From the fish’s lateral line, to the complex nature of the dolphin’s inner pinna, there is a trend towards localization using higher and higher frequencies. For mammals, higher frequency hearing was primarily driven by the need to localize, but it’s clear there are other evolutionary factors at play. We’ve only taken a look at hearing on a relatively isolated path, but it would be interesting to put it into a broader evolutionary context, considering all facets of the organism's life, how different senses play together, especially hearing and vision.

Humans are notoriously bad at listening to animals, but, as Bernie Krauss (2013) emphasizes, there is information all around us that is being used and transmitted and we don’t even know about it. Listening to an ecosystem can unlock this information. But knowing how to listen is the first step. This paper’s purpose is to illustrate how animals listen, so that we can place ourselves in their shoes to discover this hidden information.

References

Anthwal, N., Joshi, L., & Tucker, A. S. (2013). Evolution of the mammalian middle ear and jaw: Adaptations and novel structures. Journal of Anatomy, 222(1), 147–160. https://doi.org/10.1111/j.1469-7580.2012.01526.x

Clark, C. W. (2004). Baleen whale infrasonic sounds: Natural variability and function. The Journal of the Acoustical Society of America, 115(5), 2554–2554. https://doi.org/10.1121/1.4783845

Dooling, R. J., Lohr, B., & Dent, M. L. (2000). Hearing in birds and reptiles. In R. J. Dooling, R. R. Fay, & A. N. Popper (Eds.), Comparative Hearing: Birds and Reptiles (pp. 308–359). Springer. https://doi.org/10.1007/978-1-4612-1182-2_7

Garstang, M. (2004). Long-distance, low-frequency elephant communication. Journal of Comparative Physiology A, 190(10), 791–805. https://doi.org/10.1007/s00359-004-0553-0

Grothe, B. (2018). How the barn owl computes auditory space. Trends in Neurosciences, 41(3), 115–117. https://doi.org/10.1016/j.tins.2018.01.004

Grothe, B., & Pecka, M. (2014). The natural history of sound localization in mammals – a story of neuronal inhibition. Frontiers in Neural Circuits, 8. https://www.frontiersin.org/articles/10.3389/fncir.2014.00116

Heffner, H. E., & Heffner, R. S. (2018). The evolution of mammalian hearing. AIP Conference Proceedings, 1965(1), 130001. https://doi.org/10.1063/1.5038516

Heffner, R., & Heffner, H. (1980). Hearing in the elephant (elephas maximus). Science, 208(4443), 518–520. https://doi.org/10.1126/science.7367876

Krause, B. (2013). The great animal orchestra: Finding the origins of music in the world’s wild places (First Back Bay paperback edition). Back Bay Books Little Brown, Back Bay Books.

Mason, M. J., Cornwall, H. L., & Smith, E. S. J. (2016). Ear structures of the naked mole-rat, heterocephalus glaber, and its relatives(Rodentia: Bathyergidae). PLOS ONE, 11(12), e0167079. https://doi.org/10.1371/journal.pone.0167079

Nachtigall, P. E. (2016, November 22). Biosonar and sound localization in dolphins. Oxford Research Encyclopedia of Neuroscience. https://doi.org/10.1093/acrefore/9780190264086.013.103

Nachtigall, P. E., Lemonds, D. W., & Roitblat, H. L. (2000). Psychoacoustic studies of dolphin and whale hearing. In W. W. L. Au, R. R. Fay, & A. N. Popper (Eds.), Hearing by Whales and Dolphins (pp. 330–363). Springer. https://doi.org/10.1007/978-1-4612-1150-1_8

Nummela, S. (2009). Hearing. In W. F. Perrin, B. Würsig, & J. G. M. Thewissen (Eds.), Encyclopedia of Marine Mammals (Second Edition) (pp. 553–562). Academic Press. https://doi.org/10.1016/B978-0-12-373553-9.00129-2

Putland, R. L., Montgomery, J. C., & Radford, C. A. (2019). Ecology of fish hearing. Journal of Fish Biology, 95(1), 39–52. https://doi.org/10.1111/jfb.13867

Reynolds, J. E., & Rommel, S. A. (Eds.). (1999). Biology of marine mammals. Smithsonian Institution Press.

Reynolds, R. P., Kinard, W. L., Degraff, J. J., Leverage, N., & Norton, J. N. (2010). Noise in a laboratory animal facility from the human and mouse perspectives. Journal of the American Association for Laboratory Animal Science : JAALAS, 49(5), 592–597. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2949429/

Schnyder, H. A., Vanderelst, D., Bartenstein, S., Firzlaff, U., & Luksch, H. (2014). The avian head induces cues for sound localization in elevation. PLOS ONE, 9(11), e112178. https://doi.org/10.1371/journal.pone.0112178

Sound reception—Echolocation in bats | Britannica. (n.d.). Retrieved April 17, 2023, from https://www.britannica.com/science/sound-reception/Echolocation-in-bats

The ossicles and their function. (n.d.). Retrieved April 17, 2023, from http://hyperphysics.phy-astr.gsu.edu/hbase/Sound/oss.html